Smith.] 



Some Aspects of Erosion. 



'75 



fore more uniform. That is, the dissection is not so great as in 

 regions without vegetation. 



If vegetation decreases erosion on a given slope, less material 

 for cutting will be furnished to the stream at the foot of the slope- 

 Whatever effect this may have on the cutting of the stream, a more 

 important factor is the conservation of moisture in the soil, due to 

 the vegetation, and the consequent decrease in the immediate 

 amount of water given to the stream. This loss to the stream is 

 partly permanent, the moisture being taken up by the plants and 

 returned to the atmosphere, and partly only temporary, as most of 

 the water not so taken up will be given up slowly to the stream. 

 This retention of moisture tends to a smaller increase of the stream's 

 volume just after a rain; and as it is the volume of water at the 

 season of flood which is most effective in cutting, this influence is 

 an important one in reducing the rate of the stream's corrasion. 

 Widespread vegetation will, therefore, tend to decrease the rate o' 

 general erosion throughout a drainage system ; while, other things 

 being equal, the angle of the slopes covered with vegetation will be 

 greater than that of any slopes not so covered. 



CONCLUSION. 



It has been already indicated that the explanation here given of 

 planation forms is not intended to do away with the theory of 

 peneplanation. Some of the forms described as peneplains must 

 undoubtedly be accepted as such, in the established sense of the 

 word ; but it has, I think, been sufficiently shown that this hypoth- 

 esis is one which should be employed with the utmost caution- 

 For the many so-called peneplains which strain the theory beyond 

 its reasonable limits, the hypothesis of " beveling," as proposed by 

 Tarr, and modified by the present writer, is offered as an adequate 

 and much more probable explanation. 



The occurrence of a peneplain can not be predicated upon 

 evidence of a "nearly featureless plain," alone. This must be 

 correlated with all other available evidence before the cause of 

 the planation can be assigned. If one were to find widespread 

 truncation of the hilltops in a region of disturbed rocks, with 

 adjusted stream courses, in addition to evidence of the "nearly 



